Tutorial: Spark Advance

[bluetorp]

We've talked about fueling when it comes to tuning, so now let's talk about the other main component: spark advance. With spark advance, tuning is a little less concrete, insofar as there is no sensor or parameter to log to tell us that our settings are correct. However, if we understand the combustion process, it's easy to understand how advancing or retarding the spark advance effects power, and thus what's involved in tuning spark advance for maximum performance.

Let's consider the cylinder events in a four stroke-engine. First, we have the INTAKE STROKE. During the intake stroke, the intake valve is open and the piston is travelling down the cylinder. This action draws the air/fuel mixture into the cylinder. Next comes the COMPRESSION STROKE. During the compression stroke, both intake and exhaust valves are closed and the piston is travelling up the cylinder. This compresses the air/fuel mixture that was drawn into the cylinder during the intake stroke. Next we have the POWER STROKE. During this stroke, the compressed air/fuel mixture is ignitied by the spark plug, creating an explosion that drives the piston down the cylinder. Finally, we have the EXHAUST STROKE, where the exhaust valve is open, and the piston is travelling up the cylinder, thereby forcing the exhuast gasses from the power stroke out of the cylinder and into the exhaust. After this stroke, we return to the intake stroke. These four strokes are what gives the four-stroke engine its name. The non-power strokes of a particular cylinder are driven by the power stokes of the other cylinders in the engine. We use the battery to drive the starter to turn the crankshaft to begin the process, and once combustion has taken place and is driving the pistons, we no longer need the starter and the internal combustion process takes over in totality.

During the compression stroke, when the piston has achieved its maximal point of upward travel in the cylinder, we refer to this as "top dead center", or TDC. Imediately after TDC, we want to maximize cylinder pressure to drive the piston down the cylinder, and we do this by igniting the air/fuel mixture with the spark plug. In a perfect world, we could just fire the plug just after TDC has been reached. Unfortunatley, in reality, it takes time for the plug to fire, for the air/fuel mixture to ignite, for the flame to travel, and for maximum cylinder pressure to be reached from the explosion. Because of this, we need to ignite the mixture BEFORE TDC if we want to achieve maximum cylinder pressure immediately following TDC- and thus we have spark advance.

When we refer to spark advance, we refer to it in degrees BEFORE TDC. So, if our spark advance is 20*, this means that we are firing our spark plug 20* of crankshaft rotation prior to the piston reaching TDC. As our RPM increases, our piston is travelling faster and faster within the cylinder. The faster our pistion is travelling, the sooner we have to ignite the air/fuel mixture to achieve maximal cylinder pressure immediately following TDC. Understanding this, it's easy to see that as our RPM increases, so must our spark advance increase so that we can achieve maximum power during the power stroke.

With spark advance, we can have both too much and too little to achieve maximum power. If we have to little advance, we ignite the mixture too late and achieve maximum cylinder pressure when the piston is already travelling down the cylinder. This equates to a loss of power. If we have too much advance, we ignite the mixture too soon, and reach maximal cylinder pressure while the piston is still travelling up the cylinder during the compression stroke. This also equates to a loss of power. The task now becomes determining what is the optimum amount of advance to achieve maximum power for the particular RPM/load cell the engine is currently operating in.

Now, let's consider what sets our spark advance in the PCM. To begin with, we have both a "high octane" and "low octane" spark advance table, which serves as the base map of our spark advance. These tables are plotted out along the axes of RPM vs. load (in airflow). In addition to these base tables, we also have other modifier tables, such as IAT spark advance, ECT spark advance, and AFR spark advance. There are more besides these, but they are applied under specific operating conditions and are outside the scope of our current discussion. Under normal operation, the PCM utilizes the high octane table. If a certain knock retard (KR) threshold is reached, it drops down to the low octane table as a safety precaution. It is important to understand that the high/low octane tables have nothing to do with octane, and everything to do with detonation, despite how they are named.

This is the stock high octane table in a 2010 Camaro SS PCM:



This brings to bear our second consideration when it comes to spark advance: DETONATION. If we advance the spark too much, we get detonation. Sometimes, the amount of spark advance that gives us the most power results in detonation, especially if we do not have access to high octane fuel. Within these two considerations, we have the guidelines for setting our spark advance: we want to set it for maximum power, but we want to limit our advance so that we don't get detonation (which will manifest itself as KR). Detonation will be a factor of both our temperature (IAT and ECT) and the octane rating of our fuel.

But, how do we tune spark advance? We can think of it as being a two-fold process- part-throttle advance, and PE (performance enrichment- essentially wide open throttle, or WOT) advance.

For non-PE:

A very common (and very incorrect) method of tuning the spark advance table is to just add advance across the board, and subtract for cells that exhibit KR. This is wrong because we can advance timing to the point that we are losing power, yet still not be so advanced that we are getting detonation. The "correct" way to tune this is on a loaded dyno (such as a Mustang dyno- not a Dynojet). A loaded dyno allows the vehicle to be locked into an RPM/load cell, at which point you can manipulate the spark advance in that cell for maximum power. It can be a time consuming process, but yields the best results. Not everyone has access to a loaded dyno, or can spend that amount of time on the dyno tuning the spark advance. In those cases, it's generally a good idea to start with the stock table, and modify the advance based on the old "butt-dyno". That can require a lot of modification and driving and logging, but if you don't have access to a loaded dyno it's your best option. You'll want to make sure that you're datalogging during this process. If you're getting KR for a particular cell, you're advancing the spark too much and will want to back off a few degrees at a time until the KR goes away.

For PE:

PE spark advance tuning as a bit easier. You can do this on any dyno. Do WOT runs, and tune spark advance for max power and no detonation. One very common practice (but not necessarily a good one) is illustrated below:



In this example, the entire PE range has been set to one value. It's quick and easy (and very common), but not optimal. At lower RPMs, we need less spark advance for max power because the piston is not travelling as quickly in the cylinder. The dyno will dictate the proper amount of spark advance, but in general, max power advance will ramp up from low RPMs to torque peak, and from there will hold (or advance slightly) to power peak, unless KR is encountered.

But, what about those adder tables?

AFR spark is generally zeroed out during tuning. It adds advance during PE mode, but it's generally easier to just set PE advance straight out in the base spark tables. You won't be in the PE load cells outside of PE mode anyway.

IAT and ECT spark adder tables are also sometimes zeroed out, but I like to maintain them, ESPECIALLY in FI applications. In my opinion, a "good" tune will maintain the functionality of the IAT and ECT spark adder tables (though not necessarily at the stock values).

[bmorecam]

"A very common (and very incorrect) method of tuning the spark advance table is to just add advance across the board, and subtract for cells that exhibit KR. This is wrong because we can advance timing to the point that we are losing power, yet still not be so advanced that we are getting detonation".

I couldn't agree more. Most mail order tunes, or 1 tune for every car are mostly done in this exact form and sometimes even worst.

Let me STFU before I say too much

[bluetorp]

Let me just add a disclaimer that there are more modifiers and a lot of nuance involved that are not covered in my post above. Take it as a broad-strokes explanation.

[Nessal]

Originally Posted by bmorecam

"A very common (and very incorrect) method of tuning the spark advance table is to just add advance across the board, and subtract for cells that exhibit KR. This is wrong because we can advance timing to the point that we are losing power, yet still not be so advanced that we are getting detonation".

I couldn't agree more. Most mail order tunes, or 1 tune for every car are mostly done in this exact form and sometimes even worst.

Let me STFU before I say too much

Man that gets me worried about mail ordered tunes.

[mojatto]

the table for the timing tune for non-PE or PE table is the same?

In VCM Editor (ECM) 12600 Main Spark vs Airmass vs RPM Open Throttle, High Octane ?